Advances in Radiation Protection in Medicine · medicine procedures used in cardiology and other...

April 16-17, 2007

Crystal Forum

Crystal City Marriott

1999 Jefferson Davis Highway

Arlington, Virginia

Forty-ThirdAnnual Meeting Program

Advances in RadiationProtection in Medicine

Co-sponsors

American College of RadiologyReston, Virginia

Xoran TechnologiesAnn Arbor, Michigan

During the past two decades remarkableprogress has been made in the developmentand application of new medical technologiesthat utilize radiation for the early detectionand effective treatment of cancer and otherdiseases. These advances, however, areaccompanied by many questions about howto maximize medical benefits to patients,while controlling and reducing their risksfrom exposure to ionizing radiation. Theseissues are the theme of the 2007 NCRPAnnual Meeting.

Although the many advances in medicalradiation technology have represented sig-nificant gains in the prognosis for early dis-ease detection and therapy, there are issuesregarding the safety of these new radiationmodalities that are of current interest andconcern to the medical community. Amongthese are the administration of higher radia-tion doses to patients from imaging modali-ties such as computed tomography thanfrom conventional radiography. Similarly,combined modality imaging and nuclearmedicine procedures used in cardiology andother diagnostic procedures are associatedwith relatively high patient doses. In addi-tion, the increased use of image-guidedinterventional therapeutic procedures has

increased the radiation exposure of bothpatients and medical practitioners. Specialconcerns have been raised regarding use ofthe newer radiation modalities in pediatricradiology and in imaging and radiotherapyprocedures with pregnant women.

NCRP’s 2007 Annual Meeting features presentations by physicians, medical physi-cists, and experts in radiation health effectswho will discuss the rapid growth in use ofrelatively new medical radiation diagnosticand therapeutic procedures, and the currentstate of understanding of radiation dosesreceived by patients and the associatedhealth risks. Topical areas of focus at themeeting will include diagnostic radiology,nuclear medicine, interventional radiology,radiation oncology, and interdisciplinaryissues such as the implications of radiationdose-response models for the prediction oflong-term patient responses to irradiationfrom diagnostic and therapeutic procedures.

The 2007 meeting is the third in a series ofNCRP Annual Meetings on the subject ofradiation protection in medicine. The firsttwo meetings were held in 1992 and 1999,and the proceedings can be obtained at thewebsite http://NCRPpublications.org.

Introduction

3

Advances in Radiation Protection in MedicineForty-Third Annual Meeting of the National Council onRadiation Protection and Measurements (NCRP)

Monday, April 16, 2007

Opening Session8:00 am Welcome

Thomas S. Tenforde, PresidentNational Council on Radiation Protection and Measurements

8:15 am Fourth Annual Warren K. Sinclair Keynote AddressUse and Misuse of Radiation in MedicineJames A. BrinkYale University

Diagnostic Radiology ICynthia C. Cardwell, Session Chair

9:15 am Magnitude of Radiation Uses and Doses in the United States: NCRP Scientific Committee 6-2 Analysis of Medical ExposuresFred A. Mettler, Jr.New Mexico Federal Regional Medical Center

9:40 am Dose in Computed Tomography: How to Quantitate, How to ReduceCynthia H. McColloughMayo Clinic

10:05 am Break

10:35 am Pediatric Dose Reduction in Computed TomographyDonald P. FrushDuke University Health Systems

11:00 am Diagnostic Reference Levels for Medical Imaging with Ionizing Radiation: ICRP GuidanceMarvin RosensteinICRP Committee 3 (Protection in Medicine)

11:25 am Capturing Patient Doses from Fluoroscopically-Based Diagnostic and Interventional SystemsStephen BalterColumbia University Medical Center

11:50 am Lunch

Interdisciplinary IssuesLinda A. Kroger, Session Chair

1:30 pm Update on Linear Nonthreshold Dose-Response Model and Implications for Diagnostic Radiology ProceduresRobert L. UllrichColorado State University

David J. BrennerColumbia University

1:55 pm Research Involving Human SubjectsRichard L. MorinMayo Clinic

2:20 pm Radiation and PregnancyClaire CousinsCambridge University, UK

2:45 pm Break

Nuclear MedicineEdwin M. Leidholdt, Session Chair

3:10 pm Operational Radiation Safety for PET, PET/CT, and Cyclotron FacilitiesPat ZanzonicoMemorial Sloan-Kettering Cancer Center

3:35 pm Combined Imaging Modalities: PET/CT and SPECT/CTAlan H. MaurerTemple University School of Medicine

4:00 pm PANEL DISCUSSIONJulie E.K. Timins, Moderator

Program Summary

5

4:40 pm Break

Thirty-First Lauriston S.Taylor Lecture on Radiation Protectionand Measurements

5:00 pm Introduction of the LecturerRaymond Guilmette

The Quest for Therapeutic Actinide ChelatorsPatricia W. DurbinLawrence Berkeley National Laboratory

6:00 pm Reception in Honor of the Lecturer

Tuesday, April 17, 2007

8:00 am Business Session

9:00 am Break

Diagnostic Radiology IIThomas Ohlhaber, Session Chair

9:30 am Exposure Reduction Through Quality Assurance for Diagnostic X-Ray ProceduresJill A. LipotiNew Jersey Department of Environmental Protection

9:55 am State of Art: Computed Radiographyand Digital RadiographyJ. Anthony SeibertUniversity of California Davis Medical Center

10:20 am Developments in MammographyMartin J. YaffeSunnybrook Health Sciences Centre, University of Toronto

10:45 am Trends in Utilization and Collective Doses from Medical ProceduresMythreyi BhargavanAmerican College of Radiology

11:10 am Cone-Beam Imaging in DentistryStuart C. WhiteUniversity of California, Los Angeles

11:35 am Lunch

Interventional ProceduresCharles E. Chambers, Session Chair

1:00 pm Overview of Contemporary Interventional ProceduresDonald L. MillerNational Naval Medical Center

1:25 pm Patient and Personnel Safety in Interventional Fluoroscopy ProceduresLouis K. WagnerUniversity of Texas

1:50 pm Technical Advances of Interventional Fluoroscopy and Flat-Panel Image ReceptorPei-Jan P. LinBeth Israel Deaconess Medical Center

2:15 pm Break

6

Radiation OncologyTheodore L. Phillips, Session Chair

2:45 pm New Technologies in Radiation Therapy: Ensuring Patient Safety, Radiation Safety, and Regulatory Issues in Radiation Oncology Howard L. AmolsMemorial Sloan-Kettering Cancer Center

3:10 pm Dose to Normal Tissues Outside the Radiation Therapy Patient’s Treated Volume: A Review of Different Radiation Therapy TechniquesJames A. PurdyUniversity of California Davis Medical Center

3:35 pm Patient Susceptibility to Radiation-Induced Cancer and Second Cancers Following Radiotherapy ProceduresJames M. AllanUniversity of York, UK

4:00 pm Panel DiscussionStephanie K. Carlson, Moderator

4:40 pm Closing RemarksThomas S. Tenforde, PresidentNational Council on Radiation Protection and Measurements

7

Abstracts of Presentations

8

While radiation is used in many branches ofmedicine for worthwhile diagnostic and thera-peutic purposes, the potential for misuseseems greatest in diagnostic imaging. Andamong imaging tests that use ionizing radia-tion, the potential impact of misuse is greatestwith computed tomography (CT).

“I am an adult and a physician! I don’t needyour approval for CT scans that are necessaryfor my patients!” Such statements reflect thegrowing frustration among healthcare profes-sionals who struggle with appropriate utiliza-tion of medical imaging tests that use relativelyhigh doses of ionizing radiation. In an erafocused on “pay for performance,” it is easy to focus on the radiation dose associated witha particular examination. There are numeroustechnical factors that may be manipulated,modulated or filtered to produce a dose that is as low as reasonably achievable. However,appropriate utilization of these tests is a moredifficult issue to address. In our own hospital,the physician responsible for this quote ischarged with improving the quality of ouremergency services by maximizing throughputand minimizing length of stay. Having carteblanche access to imaging tests is viewed as a quality enhancer, owing to the time saved by not having to engage in a discussion aboutthe risks versus benefits of a CT scan in a

particular patient. However, by eliminating theneed for this consultation, the responsibility ofthe radiologist as the “keeper of the keys” topotentially harmful medical imaging is elimi-nated. As a result, utilization soars and diag-nostic yield plummets. In addition to the poten-tially harmful effects on individual patients,technical and professional imaging resourcesare strained by the added work burden, andpatients with appropriate medical indicationsmay be underserved owing to the high volumeof relatively unnecessary imaging studies thatmust be performed.

The potential benefit that comes with medicalimaging in patients with known diagnoses mustbe weighed against the risks of ionizing radia-tion, taking into account the patient’s age, gen-der and body part to be examined. In most pri-mary clinical circumstances, the benefits out-weigh the risks, particularly given the potentialfor diagnoses yet unfound. However, the serialevaluation of known clinical conditions for inter-val change may represent “low hanging fruit” inthe war on over-utilization of potentially harmfulimaging tests. Intensive educational effortsmust be directed at the medical community atlarge to inspire a change in diagnostic algo-rithms to include one set of imaging tests forprimary diagnosis and another for follow-up ofknown pathology. Such a culture change must

Monday, April 16, 2007

Opening Session8:00 am Welcome

Thomas S. Tenforde, PresidentNational Council on Radiation Protection and Measurements

8:15 am Fourth Annual Warren K. SinclairKeynote AddressUse and Misuse of Radiation in MedicineJames A. BrinkYale University

Monday, April 16 (continued)

9

extend from the most senior healthcare admin-istrator to the most junior healthcare profes-sional who is charged with acquiring the necessary imaging tests.

The use of ionizing radiation in medical imag-ing is extending rapidly beyond evaluation ofpatients with known or suspected diagnosesto include several screening applications.While screening mammography was the onlysuch application in use for several decades,we have seen a rapid emergence of screening

CT applications in the colon, heart and lungs.Each of these tests are proposed for patientswith risk factors for a particular diagnosis andno signs or symptoms. Most analyses to-datefocus on the cost of screening with such testsand do not factor in the risk of a fatal cancerfrom the related radiation exposure. Both mustbe considered relative to the benefit of detect-ing the diagnosis during its preclinical phaseand potentially curing it before it becomeslethal.

NCRP Scientific Committee 6-2 (SC 6-2) iscurrently working to estimate the radiationexposure of the U.S. population from allsources and will produce an NCRP report in2008. One subcommittee is specifically evalu-ating medical patient exposures. The last com-prehensive evaluation regarding the types ofmedical radiation procedures, their magnitude,and annual per capita effective doses wasdone more than two decades ago.

The medical subcommittee has examined avariety of data sources, including commercialsurveys, Medicare, U.S. Department ofVeterans Affairs, and insurance carrier data.The data sources are primarily from 2004 and2005. These data files are the most compre-hensive for diagnostic and nuclear medicineexaminations, and less complete for interven-tional procedures and radiation therapy. Thisinformation has provided a realistic estimate ofthe number and types of examinations beingdone, as well as the breakdown by broad agegroups. The subcommittee also has collectedand analyzed data on the absorbed dose,

computed tomography (CT) dose index, andother parameters necessary to estimate effec-tive dose per procedure and ultimately, collec-tive dose to the U.S. population. An issue thatremains is the most appropriate values of radi-ation weighting factors to be used in estimat-ing effective doses for diagnostic x-ray andnuclear medicine examinations.

What has become clear from this study is thatmedical exposures have increased rapidly overthe past two decades, not only in number butalso in dose. The largest increase has comefrom increased use of CT scanning proce-dures, which have increased 10 to 15 % annu-ally while the U.S. population has increased at<1 % per year. There were about three millionCT scans performed in the United States in1980, and this number has grown to about 60million CT scans in 2005 (an average of aboutone scan for every five persons). Much of theincrease has come from an increasing numberof CT machines, newer and faster technology,and new clinical uses of CT such as the evalu-ation of pulmonary emboli, lung nodules, and

Diagnostic Radiology ICynthia C. Cardwell, Session Chair

9:15 am Magnitude of Radiation Uses and Doses in the United States: NCRP Scientific Committee 6-2 Analysis of Medical Exposures Fred A. Mettler, Jr.New Mexico Federal Regional Medical Center


10

abdominal pain. Assuming a radiation weight-ing factor of one, the effective doses from CTscans range from 1 to 10 mSv per exam, andmany patients have more than one examina-tion. Collective effective doses from CT areestimated to be in the range of 300,000 person-Sv annually.

Another large and rapidly growing source ofpatient exposures is from cardiac nuclear med-icine studies, with an effective dose of about10 mSv per examination. There are approxi-mately 20 million nuclear medicine scans doneannually in the United States, of which abouttwo-thirds are cardiac studies. The collectivedose from nuclear medicine procedures on anannual basis is estimated to be about 220,000person-Sv. It is interesting to compare thesemedical doses with the global collective dosefrom the Chernobyl accident of about 600,000person-Sv.

Currently, it appears that the increasing use ofmedical radiation technology is likely to resultin per capita annual doses close to, or greater

than, the natural background exposure level inthe United States. However, it is important tobear in mind that substantial clinical benefitsoften result from exposures associated withdiagnostic and therapeutic medical radiationprocedures. It should, however, be noted thatage and illness of the medical population is not taken into account with effective dose calculations.

The SC 6-2 subcommittee is also addressingpotential increases in the use of radiation inmedicine, and the doses to which patientshave been exposed since 2005 and to whichthey are likely to be exposed in the near future.Areas of interest include, among other expo-sures associated with the introduction of digitalfilmless radiology systems, 64-slice CT scan-ners, combined positron emission tomographyand CT scanners, combined single photonpositron emission tomography and CT scan-ners, and CT screening for coronary arterystenosis and calcification.

9:40 am Dose in Computed Tomography: How to Quantitate, How to ReduceCynthia H. McColloughMayo Clinic

The fundamental radiation dose parameter incomputed tomography (CT) is the CT doseindex (CTDI). CTDI represents the integralunder the radiation dose profile of a single-axial scan, estimates the average dose from amultiple-scan examination, and is a directlymeasurable and standardized quantity. CTDIvol

is a radiation dose parameter defined by theInternational Electrotechnical Commission thatprovides a single-dose parameter, based on adirectly and easily measured quantity, whichrepresents the dose within the scan volume toa standardized phantom. All current CT scan-ners display the value for CTDIvol on their console. This feature can allow the clinician tocompare the radiation output from differentimaging protocols. CTDIvol is expressed in theunit of milligray (mGy). Dose-length product

[DLP (mGy cm–1)] is derived from the productof the scan length (cm) and CTDIvol.

The parameter of greatest interest in assessingand comparing radiation doses and biologicrisk is the effective dose. It is calculated from organ dose estimates using weighting coefficients prescribed by the InternationalCommission on Radiological Protection, whichhave evolved over time. It is a single-doseparameter that reflects the risk of a nonuniformexposure in terms of a whole-body exposure.Effective dose is expressed in the unit of millisievert (mSv).

To manage the dose from CT while maintainingdiagnostic image quality, scanner manufac-turers have implemented several technical features, including more aggressive beam


11

filtration, tube current (milliampere) modulationschemes, noise-reducing image filters, andspecialized pediatric protocols. Modulation ofthe tube current is an effective method ofmanaging the dose. However, the distinctionsbetween the various tube current modulationproducts are not clear from the product namesor descriptions. Depending on the scannermodel, the tube current may be modulatedaccording to patient attenuation or a sinu-soidal-type function. The modulation may befully preprogrammed, implemented in near-realtime by using a feedback mechanism, orachieved with both preprogramming and afeedback loop. The dose modulation may

occur angularly around the patient, along the long axis of the patient, or both. Finally, thesystem may allow use of one of several algo-rithms to automatically adjust the current toachieve the desired image quality. Modulationboth angularly around the patient and alongthe z-axis is optimal, but the tube current mustbe appropriately adapted to patient size fordiagnostic image quality to be achieved. Dosereductions of 20 to 40 % have been reportedusing milliampere modulation schemes. In cardiac CT, even more aggressive dose reduc-tions can be achieved by reducing the tubecurrent during specific portions of the cardiaccycle.

10:05 am Break

10:35 am Pediatric Dose Reduction in Computed TomographyDonald P. FrushDuke University Health Systems

Patient safety is a central issue in medicalimaging and radiation protection continues tobe a key component in a safety program. Thebalance between radiation dose and imagequality should be the perspective whenaddressing the issue of radiation protection.Discussing the balance between dose andimage quality in pediatric computed tomogra-phy (CT) is important for several reasons. First,

the use of all CT, including pediatric CT, isincreasing and techniques for CT in childrenmay be relatively unfamiliar. Second, there areadditional considerations for radiation doseassessment and risk in children compared withadults. Finally, there are unique aspects whenaddressing pediatric CT quality. The discussionof pediatric CT dose and image quality is alsojustified as there is currently no regulation in

Non-CT Typical Effective Dose Valuesa CT Typical Effective Dose Valuesa

(mSv) (mSv)

Hand radiograph <0.1 Head 1 – 2Dental bitewing <0.1 Chest 5 – 7Chest radiograph 0.1 – 0.2 Abdomen 5 – 7 Mammogram 0.3 – 0.6 Pelvis 3 – 4 Lumbar spine radiograph 0.5 – 1.5 Abdomen and pelvis 8 – 11 Barium enema exam 3 – 6 Coronary artery calcium 1 – 3 Coronary angiogram (diagnostic) 5 – 10 Coronary angiography 5 – 12 Sestamibi myocardial perfusion 13 – 16 Thallium myocardial perfusion 35 – 40

aAverage U.S. annual effective dose equivalent 3.6 mSv (NCRP Report No. 93, 1987).


12

the United States for the practice of CT foradults or children.

CT provides extremely useful information andcurrent practice indicates that it is becomingthe primary modality for evaluation of a varietyof disorders in both adults and children. This isespecially evident in the emergency setting.For example, CT is replacing ultrasound in the evaluation of pediatric appendicitis.Contemporary practice is not always based on outcome, but can be driven by marketing,economics, and public opinion. Just as CT usehas outpaced justification in many settings, ithas also been difficult for the radiologist tokeep up with technologic advancements, suchas cardiac-gated CT, and automatic tube cur-rent modulation. For example, with automatictube current modulation, the appropriate levelof noise for diagnostic quality may be differentin infants and children than in adults. In addi-tion to this potential unfamiliarity with rapidlyadvancing technology, the majority of radiolo-gists have no training in pediatric imaging afterresidency. Ironically, these same individuals are responsible for the majority of pediatricimaging.

Determining the dose from multidetector arrayCT is problematic. For example, the dose-length product is a commonly used estimation

of dose, but this method is imprecise. Thedose resulting from CT can be substantial andCT provides the highest dose of all medicalimaging using ionizing radiation. We found,using a 5 y-old anthropomorphic phantom,that we could configure an exam to give aneffective dose of nearly 120 mSv (unpublisheddata). Children’s tissues and organs are moreradiosensitive (at least two times), and thepotential for cancer development is more sub-stantive given the greater number of years oflife. In addition, the dose delivered to a child is higher than that to an adult when similar settings are used.

Study quality may be different in pediatric CT.First, the spectrum of injury and illness is dif-ferent in children and the imaging features maybe unfamiliar to radiologists, especially if pedi-atric examinations are infrequent in practice.Structures are often smaller, as well. Thesefactors can translate to a need for higherimage quality, and higher radiation doses. Asmentioned above, the amount of acceptablenoise may be lower with CT in young childrenand infants.

In conclusion, an understanding of the uniqueconsiderations for the balance between imagequality and dose is critical for appropriatepediatric CT.

In International Commission on RadiologicalProtection (ICRP) Publication 60, reference levels were described as values of measuredquantities at which some specified action ordecision should be taken. One particular formof reference level, the diagnostic reference level (DRL) applies specifically to medical imag-ing with ionizing radiation (i.e., medical imagingwith x rays or through diagnostic nuclear medi-cine). Use of DRLs is a mechanism to manage

patient radiation dose to be commensuratewith the medical purpose.

DRLs have no direct linkage to the ICRPnumerical values for dose limits or dose con-straints. DRLs should be selected by profes-sional medical bodies often in conjunction withhealth and radiation protection authorities andtheir values will be specific to a country orregion. DRLs are a guide to encourage good

11:00 am Diagnostic Reference Levels for Medical Imaging with Ionizing Radiation: ICRP GuidanceMarvin RosensteinICRP Committee 3 (Protection in Medicine)


13

clinical practice. It is inappropriate to use themfor regulatory or commercial purposes.

The objective of a DRL is to help avoid radia-tion dose to the patient that does not con-tribute to the clinical purpose of a medicalimaging task. This is accomplished by com-parison between the numerical value of theDRL and the mean or other appropriate valueobserved in practice for a suitable referencegroup of patients or a suitable reference phan-tom. A reference group of patients is usuallydefined within a certain range of physicalparameters (e.g., height, weight). A DRL is notapplied to individual patients.

A DRL can be used to:

• improve a regional, national or local distribu-tion of observed results for a general med-ical imaging task, by reducing the frequencyof unjustified high or low values;

• promote attainment of a narrower range ofvalues that represent good practice for amore specific medical imaging task; or

• promote attainment of an optimum range of values for a specified medical imagingprotocol.

These uses are differentiated by the degree of specification for the clinical and technicalconditions selected for a given medical

imaging task. Appropriate local review andaction is taken when the value observed inpractice is consistently outside the selectedupper or lower level.

The guiding principles for setting a DRL are:

• regional, national or local objective is clearlydefined, including the degree of specificationof clinical and technical conditions for themedical imaging task;

• selected value of DRL is based on relevantregional, national or local data;

• quantity used for DRL can be obtained in apractical way;

• quantity used for DRL is a suitable measureof the relative change in patient tissue dosesand, therefore, of the relative change inpatient risk for the given medical imagingtask; and

• manner in which DRL is to be applied inpractice is clearly illustrated.

Authorized bodies are encouraged to set DRLsthat best meet their specific needs and thatare consistent for the regional, national or localarea to which they apply.

The content of the current draft of the new setof ICRP recommendations and related guid-ance that apply to DRLs is reviewed.

Patient dose data collected from diagnosticand interventional medical procedures hasseveral uses. These can be grouped into thecategories of patient risk supervision anddepartmental quality assurance. Risk super-vision includes evaluation of the stochasticradiation load on the population and the management of individual patients receivingdeterministic levels of radiation. Quality assur-ance includes evaluation of departmental

performance against guidance levels and theevaluation of individual systems and operatorsagainst departmental norms.

The range of imaging technologies and proce-dures is large. Fluoroscopic-based procedurescan produce high individual procedure “dose”relative to most other imaging procedures.Therefore, this presentation focuses on fluo-roscopy, including the varieties of radiography

11:25 am Capturing Patient Doses from Fluoroscopically-Based Diagnostic and Interventional SystemsStephen BalterColumbia University Medical Center


14

usually accompanying fluoroscopically-basedprocedures.

Modern fluoroscopic systems are capable ofaccumulating the total air kerma delivered to areference point during a procedure, kerma areaproduct, as well as older items such as fluoro-scopic time and technical procedural values.The two direct measurements provide a muchbetter indication of patient risk than the olderitems. In particular, fluoroscopic time shouldnot be the only dose metric used to managehigh-dose interventional procedures.

Dose collection is in a transition between manual recordings of data from an individualimaging system to more highly automatedtechnologies. The Digital Imaging and

Communications in Medicine (DICOM) standard and the DICOM-DOSE project (a jointInternational Electrotechnical Commission-DICOM initiative) have the potential of enablingcollection of complete dose data from allmodalities irrespective of storage of the associated images.

Data should be collected for all procedureswhere there is any possibility of a deterministicradiation injury. Appropriately sampled datashould be sufficient for quality assurance purposes and for estimating stochastic risk.Oversampling in these cases will increase thecosts of data management without a commen-surate improvement in the reliability of the conclusions.

Diagnostic radiology is a significant and grow-ing source of population exposure to ionizingradiation, in large part because of the rapidincrease in computed tomography (CT) imag-ing. While organ doses from CT examinationsare still relatively small, they are much higherthan for conventional radiographs, and thus itis important that the risk/benefit balance becritically examined. A linear nonthresholddose-response model (or a model in whichlow-dose cancer risks per unit dose are largerthan derived from extrapolation of higher-doserisks) would imply that there is potential cause for concern about this rapid increase in

CT-based diagnostic imaging. On the otherhand, there would be less concern if low-dosecancer risks per unit dose are less than thosederived from extrapolation of higher-dose risks.

While there is convincing epidemiological evi-dence that doses of ionizing radiation aboveabout 100 mGy may increase the risk for can-cer in adults, at lower doses even the largestepidemiological studies have insufficientpower, and so it is necessary to rely on modelsfor extrapolation of potential risks. For children,or individuals in utero, there is plausible epi-demiological evidence for increased cancer

11:50 am Lunch

Interdisciplinary IssuesLinda A. Kroger, Session Chair

1:30 pm Update on Linear Nonthreshold Dose-Response Model and Implications for Diagnostic Radiology ProceduresRobert L. UllrichColorado State University

David J. BrennerColumbia University


15

risk at lower doses, corresponding to the well-established observation that radiosensitivityincreases with decreasing age; this is of somesignificance because of the rapid increase inpediatric CT, particularly for confirming appendicitis.

Two expert reports have been publishedrecently which give diametrically opposingopinions. The Biological Effects of IonizingRadiation (BEIR) VII report, from the NationalAcademy of Sciences, concludes that, at lowdoses, as the dose is lowered, the cancer risksimply decreases proportionately (a “linearnonthreshold” model) down to arbitrarily lowdoses. By contrast, a publication of the FrenchAcademy of Sciences suggests that, at verylow doses, the risk per unit dose for ionizingradiation-induced cancer is lower than thatestablished at higher doses; they go on to sug-gest that the induced cancer risks at very lowdoses may well be effectively zero, or evennegative.

This is clearly an important issue for diagnosticradiology. The arguments revolve around thebiological processes, at the molecular, cellularand tissue levels, that are involved in radiationresponse at very low doses (below ~100 mGy),compared with higher doses. There is nodoubt that the linear (nonthreshold) approachfor extrapolating risks to low doses (which hasbeen adopted by most national and interna-tional organizations) can and should be criti-cally examined. The arguments for a linearnonthreshold model at very low doses areplausible, but rely on assumptions about singlecells primarily acting autonomously, which areunlikely to be completely correct. However, atthis time it is unknown whether deviations fromthe predictions of this linear approach will belarge or small, nor even whether they willincrease or decrease low-dose cancer riskestimates. We are only just beginning toscratch the surface of our understanding of theimpact of intercellular interactions and tissueinteractions on very low-dose cancer risks, andso it is premature at this time to be advocatingchanges in policy or practice.

1:55 pm Research Involving Human SubjectsRichard L. MorinMayo Clinic

Human subjects have been involved inresearch studies for centuries. Originally, theyliterally were subjects, often unaware that they were involved in research studies involv-ing drugs, devices, surgical techniques, orradiation exposure among others. The use ofhumans in research studies is important sinceanimal models do not always accurately pre-dict human response. However, the times havetruly changed. Currently, humans involved inresearch are not just subjects but volunteers.The regulations (both state and federal) regard-ing human use in research have progressed to protect the safety and quality of both thehuman interactions and the research studies.

The current legislated structure of institutionalreview boards (IRB) has provided the

necessary basis and review procedures forhuman-use research studies. In addition, themandatory education of principal investigatorsand coinvestigators regarding both abuse andimproper use of humans in research, in addi-tion to both local and federal regulation, hasraised considerable consciousness regardingthese issues. This has also led to increasedscrutiny regarding external funding. It will beimportant to continue to have voluntary humaninvolvement in research, mostly due to the natural variation among humans and the smalldifferences sought to be discerned.

Studies involving ionizing radiation will con-tinue to receive heightened scrutiny due to theever increasing pace of new technology devel-opment and the continued debate regarding


16

the effects of ionizing radiation at diagnosticimaging exposure levels. The IRB assessmentof relative exposure levels for diagnostic imag-ing research studies will continue to receiveclose attention. Thus, these open discussions

will continue to protect the public health andsafety, as well as ensure that modern researchtechniques are utilized to develop new strate-gies for the safe and high-quality diagnosisand management of disease.

Every year thousands of pregnant women areexposed to radiation, either as patients or asemployees working with radiation. This oftencauses anxiety largely due to lack of knowl-edge of the women themselves, but also ofthose either working with them or caring forthem. The first instinct is to avoid radiation during pregnancy, however this is not alwayspossible as a pregnant patient may need inves-tigation and treatment and an employee mayhave no option but to continue working.

It is always advisable to assume that amenor-rhoea in a regularly menstruating woman is dueto pregnancy until proven otherwise. Diagnosticor therapeutic procedures involving radiationshould be delayed until after pregnancy wher-ever possible. If a procedure is consideredmedically indicated, the benefit to the mothershould outweigh the risk to the fetus. This isthe principle of justification which adopts moreimportance in a pregnant patient. Pregnantpatients may be exposed to radiation from radi-ological examinations, nuclear medicine proce-dures, and occasionally radiotherapy treatment.

Most diagnostic procedures if performed correctly with appropriate optimization do notpose an increased risk to the fetus. The doseto the fetus is obviously increased if the pelvisor abdomen is included in the primary beam.Higher doses from therapeutic procedures orradiotherapy can cause significant fetal harm,particularly if the pelvis is irradiated. The major-ity of diagnostic nuclear medicine proceduresuse short-lived radionuclides that do not result in a large fetal dose. Some radio-nuclides (e.g., radioiodides) cross the placenta

causing a more significant risk particularly tothe fetal thyroid.

The risk to the fetus from radiation is greatestduring organogenesis and the first trimester.The risks include nervous system abnormali-ties, malformations and cancer both in child-hood and later life.

Informed consent has to be obtained from thepatient after a full discussion of risk relative tothe procedure and this is important when thepredicted dose is >1 mGy. This may be difficultin an emergency situation when the patient isunable to give consent, and in such circum-stances the family should be counseled if possible. Fetal doses <100 mGy should not beconsidered a reason for terminating pregnancybecause this is not justified on the basis ofradiation risk. At higher fetal doses, individualcircumstances have to be discussed andinformed decisions made.

Medical radiation workers are obliged to informtheir employer if they are pregnant. When apregnancy has been declared, the InternationalCommission on Radiological Protection (ICRP) recommends an equivalent dose of not >1 mSv should be applied to the fetus.This advice differs from the recommendationsin United States of a dose limit of 0.5 mSv permonth of pregnancy and 5 mSv for the entiregestational period. Depending on duties andindividual choice, a worker may continue theirjob unchanged or decide, if possible, to moveto a position of reduced or no radiation exposure.

2:20 pm Radiation and PregnancyClaire CousinsCambridge University, UK


17

As a full-time vascular and interventional radiologist, I have personal experience of twopregnancies as a medical radiation worker andimportant issues are discussed.

ICRP Publication 84 addresses the issues ofpregnancy and medical radiation. This Report

was written with the intention of educatingmedical staff involved in everyday decisionmaking and has been widely distributed. A freeslide set is available on the subject and can bedownloaded from the ICRP website.

Positron emission tomography (PET) is now anessential and cost-effective imaging modalityin clinical practice. The definitive demonstra-tion of the clinical efficacy of, and the resultingrapid growth of, reimbursable indications for18F-fluoro-deoxyglucose (FDG) PET, the prolif-eration of high-performance turn-key PET andPET/computed tomography (CT) scanners, andthe widespread availability of FDG have com-bined to propel this dramatic advance. FDG,by far the most widely used radiopharmaceuti-cal for clinical PET imaging in general andoncologic PET imaging in particular, is highlyaccurate in detecting (~90 %) and stagingtumors, monitoring of therapy response, anddifferentiation of benign from malignantlesions.

Several factors {the relatively high administeredactivities [e.g., 370 to 740 MBq (10 to 20 mCi)of FDG], the high patient throughput (up to 30 patients per day), and, in particular, theuniquely high energies (for a nuclear medicinesetting) of the 511 keV positron-electron anni-hilation gamma rays} make shielding require-ments, workflow, and other radiation protectionissues important considerations in the designof a PET or PET/CT facility. While these topicshave been addressed in various publications,

the Report of Task Group 108 of the AmericanAssociation of Physicists in Medicine [MedicalPhysics (2006) 33(3)] provides a comprehen-sive summary of shielding design and related considerations, along with illustrativecalculations.

PET is dependent on the availability of short-lived 18F (T1/2 = 110 min) primarily in the form ofFDG, either produced in-house or purchasedcommercially. PET using shorter-lived positronemitters such as 11C (20 min), 13N (10 min), and15O (2 min), on the other hand, is impracticalwithout an in-house cyclotron. Medicalcyclotrons and associated radiochemistry facilities are now fairly numerous (well over 100 worldwide) and, of course, present theirown radiation safety issues. In addition to theradioactive product, sources of exposureinclude neutrons, a common end-product of the nuclear reactions used to producepositron-emitting radionuclides, and radioac-tive activation products in the various cyclotroncomponents and surrounding concrete. A keydecision in the installation of such a facility isthe choice between an unshielded and self-shielded cyclotron. While experienced person-nel generally prefer the unshielded designbecause shielding restricts access for repair

2:45 pm Break

Nuclear MedicineEdwin M. Leidholdt, Session Chair

3:10 pm Operational Radiation Safety for PET, PET/CT, and Cyclotron FacilitiesPat ZanzonicoMemorial Sloan-Kettering Cancer Center


18

and maintenance, the popular self-shieldedconfiguration avoids the expensive and time-consuming construction of a concrete vaultand reduces ambient neutron and gamma-rayradiation levels to the point that the cyclotroncould be located within the radiochemistrylaboratory. The design of that laboratory,largely dictated by the short half-life of 18F and other positron-emitting radionuclides, is intended to provide expeditious, short-distance transport of the starting material (i.e., the cyclotron-produced radionuclide),reagents, and packaging/dispensing materials.All such laboratories nowadays include lead-lined hot cells equipped with manipulatorarms, computer-controlled radiosynthesisunits (“boxes”), and air extraction capabilities

for passing air through a charcoal filter to trapradioactive gases and volatiles before releaseto the general environment.

Published studies have shown that that theradiation doses to personnel working in PET orPET/CT facilities and in cyclotron and associ-ated radiochemistry facilities can be main-tained below, and generally well below, thepertinent regulatory limits; the highest doses,not surprisingly, are generally accrued byradiochemistry personnel. This presentationwill review the basic radiation safety aspects,including shielding, facility design, and work-flow, of these increasingly important facilities inmodern medicine.

Nuclear medicine has long been recognized for its value as a functional imaging modalitywhich provides unique information related tocellular and organ function including: bloodflow, biochemistry, and metabolism. Traditionalnuclear medicine drugs (radiopharmaceuticals)have utilized single-photon emitters for detec-tion by conventional gamma cameras andsince the early 1990s have been used fortomographic imaging [single photon emissioncomputed tomography (SPECT)]. Because oftheir chemical structure, SPECT radiopharma-ceuticals permit only limited evaluation of cer-tain metabolic processes. Positron emissiontomography (PET) radiopharmaceuticals werepreviously utilized solely in academic medicalcenters because of the need for a localcyclotron to produce these imaging agents.Their importance, however, has long been rec-ognized as they permit more advanced imag-ing of processes such as glucose metabolism,protein synthesis, gene expression, tissuehypoxia, and receptors at a cellular level.

Recently, PET imaging has rapidly beenadopted into clinical practice in community

hospitals and outpatient imaging centers ascommercial suppliers have made the PETradiopharmaceutical 18F fluoro-deoxyglucose(FDG) widely available, and PET studies havebeen approved for reimbursement for a widerange of applications. While PET cameras haveimproved anatomic resolution compared toSPECT cameras the spatial resolution of both SPECT and PET remains limited whencompared to x-ray (transmission) computedtomography (CT). CT is known to provide veryhigh-quality imaging which depicts anatomicdetail with high spatial resolution. Combinedimaging devices now integrate both SPECTand PET cameras with CT scanners into a single device (SPECT/CT or PET/CT). Thesenew imaging devices now provide both themetabolic and functional information fromSPECT or PET combined with the high spatialresolution and anatomic information of CT.Because the two sets of images are fused,areas of normal and abnormal metabolic activ-ity can be mapped to recognizable anatomicstructures. This fusion of function and anatomyhas quickly demonstrated its clinical value par-ticularly in the areas of oncology, cardiology

3:35 pm Combined Imaging Modalities: PET/CT and SPECT/CT Alan H. MaurerTemple University School of Medicine


19

and neurology. PET/CT is currently most com-monly used in the area of oncology where ithas demonstrated advantages over PET aloneor CT alone not only for diagnosis but also forinitial staging of a patient’s cancer and forassessing the patient’s response to therapyand, if needed, later restaging. PET/CT hasbecome accepted as a standard of care forjudging the effectiveness of treatment for manycancer patients. Studies have shown that whilePET and CT are complementary, the fusion ofboth modalities results in much higher diag-nostic accuracy. Based on this PET/CT experi-ence there is now increasing utilization of

SPECT/CT for other more routine nuclear medicine studies.

In this lecture, the current clinical applicationsof SPECT/CT and PET/CT fusion imaging arediscussed. This fusion of nuclear medicineimaging with CT comes with some obviousincrease in radiation exposure to patients. Inaddition to an introduction to the current tech-nologies, the methods being employed tomaximize the information from these studies,while reducing as much as possible the inher-ent radiation exposure to the patients, are alsodiscussed.

4:00 pm PANEL DISCUSSIONJulie E.K. Timins, Moderator

4:40 pm Break

Thirty-First Lauriston S. Taylor Lecture on Radiation Protection and Measurements

5:00 pm Introduction of the LecturerRaymond Guilmette

The Quest for Therapeutic Actinide ChelatorsPatricia W. DurbinLawrence Berkeley National Laboratory

6:00 pm Reception in Honor of the LecturerSponsored by


20

Traditional state x-ray inspection programsconcentrate on measurement of x-ray machineparameters such as kilovolt peak and mil-liampere, timer accuracy, collimation, etc. In 1996, the New Jersey Radiation ControlProgram began a paradigm shift from the traditional inspection to an outcome-basedinspection that concentrated on two indicatorsof performance: image quality and entranceskin exposure (ESE). Through extensive outreach and involvement of stakeholders, a new approach was designed that placed an emphasis on quality assurance. Key to thepositive outcome has been the credentialing of medical physicists.

On January 16, 2001, the final regulation entitled “Quality Assurance Programs forMedical Diagnostic X-ray Installations”(N.J.A.C. 7:28-22) was adopted. The new regulations require that each facility usingdiagnostic medical x-ray equipment (includingradiographic, fluoroscopic, x-ray bone densito-metric, and computed tomographic) mustestablish and carry out a quality assuranceprogram. The new regulation specifies thequality control tests, frequencies and stan-dards that are part of the quality assuranceprogram. To assist physicians, chiropractors,podiatrists and the radiologic technologistsemployed by them, four compliance guidance

documents were prepared: Quality AssuranceManual, Radiographic Quality Control,Fluoroscopic Quality Control, and ComputedTomography Quality Control. Five years of datahave been gathered during inspections. Bothentrance skin exposure and image quality arechecked and the inspectors conduct an auditof the facility’s quality assurance program.Entrance skin exposure has been decreasedby 34 % for lumbar spine, 46 % for chest, and66 % for foot x-ray procedures.

Criteria for image quality have been developedand tested. When the Bureau of RadiologicalHealth inspects a facility, an image of a phan-tom is taken and scored by the inspector. Sixcriteria are evaluated (background density, highcontrast resolution, noise and artifacts, densityuniformity, low contrast detail, and low con-trast resolution). When the inspection resultsare input into the computer, a report is gener-ated and sent to each facility. This reportscores each of the six tests as excellent, good,fair or poor, and provides an overall score ofthe image quality. Facilities with poor imagequality scores are asked to consult with theirphysicist, determine the cause, make changes,and send a report of their findings and correc-tive actions to the Bureau of RadiologicalHealth within 30 d. Image quality has improvedby 22 %.

Tuesday, April 17, 2007

8:00 am Business Session

9:00 am Break

Diagnostic Radiology IIThomas Ohlhaber, Session Chair

9:30 am Exposure Reduction Through Quality Assurance for Diagnostic X-Ray ProceduresJill A. LipotiNew Jersey Department of Environmental Protection

21

Tuesday, April 17 (continued)

In April 2005, quality improvement initiativeswere extended to the larger dental x-ray com-munity. Through outreach and informationsharing, stakeholders were instructed in thefactors that affect patient radiation exposureand image quality and were encouraged totake actions to improve in these areas.Information on patient ESE at dental facilities

has been collected since 2002. All registereddental facilities (5,000) have received an ESEreport comparing their results to the rest of thedental facilities in New Jersey. As of July 1,2006 the Bureau of Radiological Health beganconducting re-inspection of dental machinesbeginning with those with extremely high ESE.Results of this effort are presented.

The widespread implementation of digital radi-ography (DR) for medical imaging applicationshas increased the need to keep up with rapidlychanging technology and the paradigm shiftconfronting all users, including technologists,radiologists and physicists. DR devices fordiagnostic medical imaging can be classifiedinto two major categories: (1) cassette-based,passive detectors, chiefly the domain of photo-stimulable storage phosphor systems, alsoknown as computed radiography (CR); (2) cas-sette-less, integrated detectors using activereadout devices, which include charge-coupleddevice and thin-film transistor arrays. Theselatter systems are often categorized as “direct”or “digital” radiography (DR). Technologicaladvances are blurring the differences betweenCR and DR, as there are CR systems availablewith integrated, high speed readout, and someDR devices with a portable, cassette-basedform factor. Advanced applications made possible by high throughput, flat-panel DRdetectors are becoming an important part ofthe clinical routine and future expectations.Examples include dual-energy radiography anddigital tomosynthesis. Fully three-dimensionalcone-beam computed tomography, achievedby rotating a two-dimensional digital detectoraround the object with full volumetric recon-struction, is providing cross-sectional and volumetric views for angiography and breastimaging.

Image quality, of paramount importance forany digital detector, is largely determined by

image pre- and post-processing algorithms,requiring proper setup and tuning during initialimplementation, acceptance testing, and qual-ity control. A common misperception is that allDR devices can produce acceptable images at a lower patient dose due to internal scalingand signal adjustment compared to typicalscreen-film cassettes such as rare-earth 400-speed systems. In fact, however, some digitalsystems require as much as a twofold higherradiation dose for similar signal to noise char-acteristics because of poorer quantum detec-tion efficiency and resultant higher noise(quantum mottle and electronic noise), whileothers require the same or slightly lower dose.For all digital systems, unintentional overexpo-sure of the patient is possible without anydirect knowledge by the technologist or radiol-ogist, as the images have high signal to noiseratio and nothing apparently “wrong.” This isproblematic, either because of unavailablefeedback that overexposure has occurred, orinadequate knowledge by the user regardingthe exposure index value provided by the man-ufacturer. Technologists must be made awareof potential overexposure tendencies, and payclose attention to radiographic techniques andpatient dose. Technique charts should beposted at all operator consoles and withportable equipment. Additionally, radiologistsshould be aware of, understand, be able todetermine, and monitor the exposure index. A complicating factor is the many differentexposure index schemes reported by the various detector systems. The American

9:55 am State of the Art: Computed Radiography and Digital RadiographyJ. Anthony SeibertUniversity of California Davis Medical Center


22

Association of Physicists in Medicine is cur-rently working on a proposal to standardize theexposure index in cooperation with equipmentmanufacturers for all CR and DR devices. Thisis one of many steps that will assist in the

proper use of DR systems. Ultimately, theusers of such devices must be aware of theissues and methods for optimization of imagequality at the lowest achievable dose.

Mammography has long been established as a useful tool for finding breast cancer insuspicious areas in the breast, identified by awoman or her physician. More recently, thecontribution of screen-film mammography toreduction of mortality from breast cancer, whenused for routine screening of asymptomaticwomen over the age of 40 has been clearlydemonstrated.

Early mammography systems were relativelyprimitive in design and the direct exposure filmimage receptors were inefficient, requiringrather high doses (~20 mGy) to the breast. Inthe 1970s, 1980s and 1990s, considerabletechnical development took place to optimizeimage acquisition and display including intro-duction of intensifying screens, fixed focal-film distances, new target-filter combinations,improved breast compression, automaticexposure control, grids, better films and processing methods, and dedicated viewingsystems. These developments resulted inimages of improved diagnostic quality pro-duced at considerably lower dose. In fact,doses dropped to the point where the signal tonoise ratio and contrast of mammograms suf-fered. As the importance of adequate contrastand spatial resolution along with low noise inensuring high diagnostic quality became betterappreciated, doses gradually increased to sta-bilize at a higher level but still markedly belowdoses used in the early 1970s.

Despite these developments in screen-filmmammography, there were fundamental barri-ers related to contrast, dynamic range, detec-tor efficiency, and image viewing that limited

the performance of mammography. Digitalmammography addressed these limitations bydecoupling image acquisition, storage and dis-play, and attempting to optimize each of theseprocesses separately. Mammograms could beviewed on a computer monitor, enhanced digi-tally and easily transmitted from one locationto another. The recently-published results ofthe Digital Mammography Imaging ScreeningTrial showed that for certain groups of womendigital mammography provided greater sensi-tivity of cancer detection in screening than film.

There are still important challenges for breastcancer detection. Digital mammography is farfrom perfect and variability of performance ofinterpreters is a major factor responsible forthis. It is essential to ensure that x rays areused as efficiently as possible to produce useful diagnostic information. This can beachieved in part through improved quality control procedures and also by leveraging newbreast imaging applications on the platform of digital mammography. These include com-puter-assisted detection and diagnosis to maximize performance of the interpreter, three-dimensional techniques like tomosynthesis ordedicated breast computed tomography toimprove conspicuity of cancers by eliminatingsuperposition effects, contrast imaging toexploit functional changes occurring with cancer, and many other new techniques.

Additionally, there is the opportunity to employmodalities that provide complementary infor-mation and do not require the use of ionizingradiation such as breast magnetic resonanceimaging and ultrasound.

10:20 am Developments in MammographyMartin J. YaffeSunnybrook Health Sciences Centre, University of Toronto

23


In the end, the most effective way to findbreast cancer at an early enough stage whereit is virtually 100 % curable may be through

tracers, which are molecularly targeted to thecancer. This is an exciting area which is still inits infancy.

Estimates of collective radiation doses frommedical procedures primarily use data from twosources: volumes of procedures and dose perprocedure. This presentation will describe available data on volumes of procedures, therates at which they have grown in recent years,and how these data are used in estimating collective doses.

The rate of growth of medical procedures overall, and the dramatic growth in the volumeof imaging procedures in particular, have beenthe subject of much attention during the past 5 y. The Blue Cross-Blue Shield Medical CostReference Guide for 2006 reports a 38 %increase in the number of diagnostic imagingcenters and a 34 % increase in diagnostic imag-ing procedures between 2001 and 2004. Thegrowth in imaging has not been uniform acrossimaging modalities, sites of service, physiciantypes, or over time. For example, the utilizationof general radiography has been relatively sta-ble, but the volume of computed tomographyimaging has been growing at over 10 % peryear since 2001. Procedures in nonhospital settings have been growing much more rapidlythan inpatient procedures. The growth in thevolume of procedures by nonradiologists in nonhospital settings has been much more rapidthan the corresponding imaging by radiologists.While the volume of imaging procedures hasbeen growing throughout the last decade, therates of growth have been much higher in recent years.

This presentation will bring together recent information from a variety of sources (Medicare,private surveys of facilities, public-use surveys,and other publicly reported data) to illustrate thetrends in medical imaging and radiation therapy

procedures in the United States during the past decade. Special attention will be paid todifferences in growth rates across sites of service and imaging by “nonradiological” physicians, because these could be associatedwith wide variation in types of equipment, levelsof regulation and oversight, and knowledge andexperience related to radiation safety. There willbe a description of the distribution of imagingvolume by patient age, particularly imaging for pediatric patients versus adult patients,because these groups receive different doses.The presentation will illustrate the differences in trends across age groups and explore whichtypes of procedures grew most rapidly for each group.

Data availability is not uniform for all types of imaging; for example, there is not as muchinformation on dental imaging as there is onother medical imaging. In general, proceduresthat are not covered by major insurance payersare difficult to track down, but are small enoughin total volume as to not affect the overall findings.

Radiation therapy procedures affect a verysmall proportion of the population, and theiroverall volume has not grown as dramatically as that of medical imaging. However, there havebeen significant changes in patterns of careover time with notable implications for patientdose. The presentation will include compiledpublished information on some of these trends.

Finally, there will be discussion of some implications of the growth in diagnostic radiation utilization and collective dose for treating physicians and the need for increasedawareness and caution on their part.

10:45 am Trends in Utilization and Collective Doses from Medical ProceduresMythreyi BhargavanAmerican College of Radiology


24

Cone-beam imaging is being used in radio-therapy for positioning and treatment planning,scientifically for scanning small animals, andfor a variety of industrial applications. Duringthe last 5 y cone-beam imaging has alsogained a broad acceptance in dentistry, especially in the United States, Europe, Japanand Canada. Currently there are about 1,000machines worldwide and the number of installations is growing rapidly. Further, somemanufacturers of conventional panoramicmachines are modifying their units for cone-beam imaging.

Cone-beam machines emit an x-ray beamshaped like a cone rather than as a fan as in acomputed tomography (CT) machine. After thisbeam passes through the patient the remnantbeam is captured on an amorphous silicon flatpanel or image intensifier/charge-coupleddetector. Unlike CT, there is no post-patientcollimation. As a result the image is capturedwith few wasted photons but is degraded byscattered radiation. The beam diameter is upto 12 inches in diameter and exposes theregion of interest in one pass around thepatient. Various machines capture from 160 to599 basis images. These images are used to compute a volume from which axial, sagittalor coronal images, or planar or curved reconstructions in any arbitrary plane can beextracted. Three-dimensional images of boneor soft tissue surfaces can be generated.

In dentistry the most common indications for cone-beam imaging are assessment of the jaws for placement of dental implants,evaluation of the temporomandibular joints forosseous degenerative changes, examination of

teeth and facial structures for orthodontictreatment planning, and evaluation of the prox-imity of lower wisdom teeth to the mandibularnerve prior to extraction. These imaging needsall rely on the three-dimensional nature of theimage reconstructions. Cone-beam images areattractive in dentistry because the image qual-ity is superior to conventional tomography thatit replaced. Cone-beam images also replacepanoramic images for some of these needsbut are unlikely to soon replace conventionalintraoral periapical or bitewing images. Cone-beam images also can be displayed withoutmagnification, a feature that is particularlyimportant for placement of implants and ortho-dontic treatment planning. The main limitationsof dental cone-beam images compared toconventional CT are the lack of a soft-tissuewindow and higher image noise.

The radiation dose from cone-beam imagingdepends on the specific brand as well as theexposure factors used and can vary by a factorof 20 times. At the low end the effective doseis about 44 μSv for a large field of view. Thisvalue is less than a conventional full-mouth set of dental x-ray views, six to seven times a panoramic view, and perhaps 2 to 5 % of a conventional CT of the same region. Thecost of the equipment is relatively low, about$150,000 to $300,000. Most dental cone-beamunits are used in universities, offices of ortho-dontists, oral surgeons and periodontists, andin dental x-ray laboratories. A major issue tobe considered is the training of individualsmaking and interpreting cone-beam images,both in terms of technical operation of theunits as well as their qualifications for evaluating the whole imaged volume.

11:10 am Cone-Beam Imaging in DentistryStuart C. WhiteUniversity of California, Los Angeles

25


Interventional fluoroscopy procedures use ionizing radiation for guidance as small instru-ments such as catheters are manipulatedthrough blood vessels or other pathways in the body. As compared to open surgical proce-dures, interventional fluoroscopy proceduresrequire a very small incision and permit shorterrecovery times. As a result, these procedureshave become very common. As an example, in 2002 an estimated 657,000 percutaneoustransluminal coronary angioplasty procedureswere performed in adults in the United States.From 1996 to 2000, the rate of coronary arterystent insertions doubled from 157 to 318 per100,000 adults aged 45 to 64.

At the same time, more complex interventionalfluoroscopy procedures have been introduced.This is due to the development of new devicesand procedures, such as endografts for thetreatment of abdominal aortic aneurysms, thedevelopment of vertebroplasty, kyphoplastyand uterine artery embolization, and increasinguse of fluoroscopic guidance during complexendoscopic biliary and upper urinary tract procedures. As the complexity of these procedures has increased, radiation doses topatients and healthcare personnel have alsoincreased.

Many interventional fluoroscopy procedureshave the potential for high patient radiationdoses, and some (particularly embolizationprocedures) are typically high-dose proce-dures. Absorbed skin doses >5 Gy may occur.Because most patients are past reproductiveage and have serious underlying medical prob-lems, their life expectancy is shortened as

compared to the general population. As aresult, deterministic radiation effects, princi-pally skin injury, are usually of greater concernthan stochastic effects. Fortunately, seriousinjuries are uncommon. The majority ofreported radiation-induced skin injuries havebeen associated with coronary artery angio-plasty and stent placement, cardiac radio-frequency ablation procedures, embolization procedures, or transjugular intrahepaticportosystemic shunt creation.

The risk/benefit analysis for interventional fluo-roscopy procedures differs from the analysisfor diagnostic radiology procedures. Unlikediagnostic radiology procedures, all interven-tional fluoroscopy procedures provide a clearbenefit for the patient. In addition, the risk ofradiation-related injury is far less than that forother procedure-related complications, so therisk/benefit analysis is relatively straightfor-ward. The patient is far more likely to beinjured by catheter manipulation than by theradiation beam.

An important goal of all interventional fluo-roscopy is to achieve clinical success usingthe least amount of radiation consistent withadequate imaging guidance. However, mostinterventional procedures require high qualityimages, long fluoroscopy time, or both. It iscritically important to train operators how toachieve the maximum possible dose reductionconsistent with acceptable image quality.Simple techniques exist which can accomplishthis. These include the use of reduced-dosepulsed fluoroscopy, collimation, and dosespreading. These techniques are simple, but

11:35 am Lunch

Interventional ProceduresCharles E. Chambers, Session Chair

1:00 pm Overview of Contemporary Interventional Fluoroscopy ProceduresDonald L. MillerNational Naval Medical Center


26

they require modern, well-maintained equip-ment, operator education and motivation.

Many interventional fluoroscopy procedureswere developed by radiologists, but these procedures are now performed by a rapidlyexpanding number of healthcare providers in a wide range of medical specialties. Theseinclude cardiology, vascular surgery, neuro-surgery, pain management, orthopedic surgery,and many other medical and surgical disciplines.

Training in radiation physics, biology andsafety has long been incorporated into radiol-ogy residency programs. The cardiology andpain management medical communities haverecently recognized the need for training inradiation physics and radiation safety.Unfortunately, most other operators have littletraining in radiation science or protection

measures, and are not motivated to becometrained.

Training requirements may be mandated byprofessional societies, accreditation organi-zations such as the Joint Commission onAccreditation of Healthcare Organizations, orgovernmental regulation. In the United States,only the individual states have the authority to require a specific knowledge base prior tooperation of fluoroscopy equipment. To date,only a handful of states have mandated specific training and licensing for physicianswho perform fluoroscopy.

Physicians, technologists, medical physicists,fluoroscopy equipment manufacturers, andmedical and governmental organizations sharethe responsibility to optimize radiation dosesto patients undergoing interventional fluoroscopy.

Radiation-induced stochastic and deterministiceffects in patients and in practitioners exist.Circumstances responsible for documentedeffects provide an abundance of informationregarding practice techniques and habits thatmust be in place to prevent deterministiceffects and to appropriately limit the occur-rence of stochastic effects. Radiation manage-ment to limit risk must be balanced againstcertain factors indigenous to medical proce-dures. For example, the medical benefit of aprocedure must be considered in an appropri-ate manner relative to the overall risk, of whichradiation represents only one agent of concern.For practitioners, the regard for radiation safetymust be considered in light of the risks thatcertain radiation-protection practices pose tothe practitioner. An example of this is the con-sideration of the protection provided by a leadapron versus the weight of that lead apron andthe ergonomic considerations associated withthat weight. The risk of injury to the spine from

a heavy lead apron is as important a consider-ation as radiation-induced disease.

Risks to patients from complex fluoroscopi-cally-guided procedures are associated withlong fluoroscopy times, irradiation throughthick body parts, and no monitoring of dose tothe patient, among other things. Despite thesefacts, in facilities where injuries have occurredfew had initiated any actions in response to the 1994 advisory of the U.S. Food and DrugAdministration (FDA) about the means to avoidthem. Many had disregarded the warningbecause fluoroscopically-induced radiationinjury was rare and the FDA advisory was notregulatory.

In response to the fact that dose monitoringhas previously not been readily available, theFDA now requires that manufacturers incorpo-rate dose monitoring devices into their fluoro-scopic equipment. However, simply requiringmanufacturers to provide dose information will

1:25 pm Patient and Personnel Safety in Interventional Fluoroscopy ProceduresLouis K. WagnerUniversity of Texas


27

have little benefit if physicians are not trainedin the use of such information.

Physicians are sometimes misled by manufac-turers who tout that their equipment is “lowdose.” For example, while some have claimedgreat strides in dose reduction with modernflat-panel devices, patients still have beeninjured from procedures that employ thesemachines. Further, while many modernmachines are equipped with high-powered

technology to reduce dose and dose rate tothe patient, training of users in the full scope ofdose management techniques is lacking. Thebottom line is that the Achilles heel of all dosemanagement and dose limiting devices is thetraining that the user has in employing them.This presentation will focus on the lessonslearned from radiation injuries and will try toidentify shortfalls in the methods so far pro-moted to limit radiation risk in medicine.

During the past decade, interventional fluoro-scopic systems equipped with image intensi-fiers have benefitted from technical advancesin x-ray tube, x-ray generator, and spectral-shaping filter technologies. While thephotoconductor (or phosphor plate) x-raydetectors and signal capture thin-film transistorarrays and charge-coupled devices are analogin nature, not until the advent of flat-panelimage receptors would fluoroscopy become a totally digital process throughout the entireimaging chain.

The high heat capacity x-ray tube, themedium-frequency inverter type generator withhigh performance switching capability, and thepatient dose reduction spectral-shaping filterhad already been implemented on image-intensified fluoroscopy systems. These threeunderlying technologies were tied togetherthrough the automatic “image quality” controllogic so that patients receiving cardiovascularangiography procedures can benefit from“lower patient dose” with “high image quality.”

The flat-panel image receptor streamlined theimage processing due to its “digital” nature,and eliminated the need to perform analog-to-digital conversion at the point of image acqui-sition. While the changeover from image-inten-sified fluoroscopy system to flat-panel imagereceptor fluoroscopy system is part of the

ongoing “digitization of radiology,” the value ofthe flat-panel image receptor may have to beevaluated from various angles including, butnot limited to patient dose, image quality, andclinical application capabilities. It is believedthat the advantage of the flat-panel imagereceptor is yet to be explored fully.

For instance, the flat-panel image receptor isnot necessarily without any disadvantage ascompared to image intensifiers; the cost of the equipment is probably the most obvious.However, there is a potential of further loweringthe patient dose through a calibration processin which the flat-panel input dose rate may beset to one-half of what is being used today.Thus, further reducing the patient dose by afactor of two is not unrealistic.

In this presentation, the main thrust is tounderstand the details of the automatic “imagequality” control logic as seen from a fluoro-scopist’s point of view, and to show how the control logic “ties” three technologicaladvancements together to provide low radiation dose to the patient and yet makehigh-quality fluoroscopic images available for manipulation of catheters. A secondary purpose is to show how three-dimensionalangiography, by providing computed-tomogra-phy-like images, can result in reduction ofpatient dose indirectly. Although “rotational

1:50 pm Technical Advances of Interventional Fluoroscopy and Flat-Panel Image ReceptorPei-Jan P. LinBeth Israel Deaconess Medical Center


28

three-dimensional angiography” was also available with an image-intensified fluoroscopysystem, the flat-panel image receptor system

is able to accomplish the same task faster andwith considerable ease.

New technologies such as intensity modulatedradiation therapy (IMRT), image guided radia-tion therapy (IGRT), computer controlled linearaccelerators (LINACs), computerized recordand verify (RV) systems, electronic charts, digi-tal imaging, etc., have revolutionized radiationtherapy over the past 10 to 15 y. Quality assur-ance as historically practiced and as recom-mended in reports such as (1) ComprehensiveQA for Radiation Oncology: Report of AAPMRadiation Therapy Committee Task Group 40.[Medical Physics (1994) 21, 581–618], and (2) AAPM Radiation Therapy Committee TaskGroup 53: Quality Assurance for ClinicalRadiotherapy Treatment Planning [MedicalPhysics (1998) 25, 1773] is in many respectsobsolete and impractical. The quantity of datacreated by an IMRT treatment plan that mustbe transferred to a LINAC coupled with thecomplexity of the dose calculations make itimpossible to “hand check” a treatment plan in the traditional sense. RV systems first intro-duced 10 to 15 y ago began as computerschecking humans; did the radiation therapistset the LINAC correctly, etc? But over theyears RV has evolved into more complex systems that now actually “run” the LINACrather than merely monitor the actions ofhuman operators. RV has evolved into humanschecking computers rather than computerschecking humans. Often it means one com-puter checking another computer.

The more recent introduction of IGRT is lead-ing to more reliance on computer control ofpatient setup and even real-time correctionsfor intrafractional patient motion, with much ofthis also falling into the category of humanschecking computers. In short, the increasingcomplexity of radiation therapy technology andthe quantity of data required to define a treat-ment plan and patient treatment has made tra-ditional quality assurance virtually impossible.

Perhaps as a result we are seeing an increas-ing fraction of medical and seminal events inradiation therapy caused either by improperuse and/or understanding of new technology;communication failures between computers;corrupted, improperly created, or improperlytransferred data files; and “software bugs.” Inour experience errors in radiation therapy are,with rare exceptions, never the result of hard-ware failures anymore. The growth of inter- andintracranial radiosurgery, use of hypofractiona-tion, complexity of treatment plans, IGRT, andincreasing financial pressures to treat morepatients in less time will continue to fuel thisreliance on high technology and in particular,complex computer software.

In the areas of diagnosis, treatment simulation,tumor contouring, and treatment planning weare also witnessing an increasing reliance oncomplex, software driven multi-modality

2:15 pm Break

Radiation OncologyTheodore L. Phillips, Session Chair

2:45 pm New Technologies in Radiation Therapy: Ensuring Patient Safety, Radiation Safety, and Regulatory Issues in Radiation Oncology Howard L. AmolsMemorial Sloan-Kettering Cancer Center


29

imaging technology. Combinations of computedtomography, magnetic resonance imaging, mag-netic resonance spectroscopic imaging, singlephoton emission computed tomography, andpositron emission tomography image fusion arefast becoming commonplace for many types ofradiation therapy treatment plans. Quality assur-ance for these modalities is often beyond theexpertise of the radiation therapy physicist, andwe increasingly rely on manufacturer-suppliedimage transfer, fusion imaging, and picturearchiving and communication computer systemswith little understanding of how they work.

Clinical practitioners as well as governmentregulatory agencies are coming to the realiza-tion that quality assurance for new technolo-gies, especially computer software, is a majorchallenge. Increasing reliance on technologyfor tumor definition, contouring, and real-timecorrections of radiation delivery coupled withdecreasing treatment field margins and doseescalation pose challenges and dangers of acompletely different nature than what we havehistorically dealt with, and this has changedthe very nature of quality assurance.

Radiation therapy treatment planning anddelivery capabilities have changed dramaticallysince the introduction of three-dimensionaltreatment planning and continue to change in response to the implementation of newadvanced technologies. Powerful x-ray computed-tomography simulation and three-dimensional treatment planning systems havebeen commercially available since the early1990s and three-dimensional conformal radia-tion therapy (CRT) is now firmly in place as the standard of practice in clinics around theworld. Medical accelerator manufacturers haveemployed advanced computer technology toproduce treatment planning/delivery systemscapable of precise shaping of dose distribu-tions via computer-controlled multileaf collima-tor systems, by which the beam fluence is varied optimally to achieve the desired dosedistribution. This mode of conformal therapy is referred to as intensity modulated radiationtherapy (IMRT), and is capable of generatingprecise conformal dose distributions includingconcave isodose volumes which provide con-formal target volume coverage and avoidanceof specific sensitive normal tissue structures.The increasing use of IMRT has focused attention on the need to better account for

both intra- and interfraction spatial uncertain-ties, which has helped spur the developmentof treatment machines with integrated planar and volumetric advanced imaging capabilities,providing what is now referred to as image-guided IMRT, or simply image-guided radiationtherapy. In addition, there is a growing interestin replacing x rays with protons because of thephysical characteristics of the Bragg energy-deposition curve, which peaks at the end ofthe particle range, and eventually with evenheavier charged particles to take advantage ofthe greater density of energy deposition closeto the Bragg peak and hence larger relativebiological effect.

For all of these conformal modalities, the challenge of treatment planning is to create an arrangement of beams that delivers the prescribed dose to the target (tumor) volume,while keeping the dose to critical normal tissues low enough to minimize the risk of serious complications. Thus, it is essential thataccurate dose-volume tolerance data for theirradiated normal tissues be available alongwith accurate data for the specific conformalmodality used regarding peripheral dose orwhole-body dose (i.e., the dose the patient

3:10 pm Dose to Normal Tissues Outside the Radiation Therapy Patient’s Treated Volume: A Review of Different Radiation Therapy Techniques James A. PurdyUniversity of California Davis Medical Center


30

receives outside of the geometric confines ofthe treatment beams).

Three-dimensional CRT, IMRT and protonbeam therapy all provide improved target coverage and lower doses to surrounding normal tissues as compared to two-dimen-sional radiation therapy techniques. However,these are achieved at the expense of more volume of normal tissue receiving some doseand/or higher whole-body doses to distantnormal tissues. These higher whole-bodydoses are the result of increased x-ray leakage

radiation from longer beam-on times associ-ated with IMRT and neutron leakage radiationassociated with high-energy x-ray beams (>10 MV) and proton beams.

This presentation will review the dose distribu-tions for the various conformal radiation ther-apy techniques and the current status of avail-able data for normal tissues, and whole-bodydose. In addition, an update on current effortsin clinical trials that use these advanced tech-nologies and the reporting of volume and dosedata will be presented.

Cancer survivors are at a significantlyincreased risk of developing a second malig-nancy as a consequence of the radiotherapyused to treat their primary malignancy. Such is the problem that second malignancies areone of the leading causes of death in long-term survivors of Hodgkin’s lymphoma. Muchresearch has focused on elucidating the relationship between radiation dose and site-specific cancer risk, and how this relationshipis affected by host factors such as age, gender, co-morbidities, and exposure to otherpotential carcinogens.

By contrast, there is a relative paucity of data on host genetic susceptibility to secondprimary cancers following radiation exposure.Animal model systems suggest a stronggenetic basis underlying susceptibility to radiogenic cancer. In humans, research hasfocused on investigating loci with relatively rare putative high-penetrance risk alleles, suchas ataxia telangiectasia (ATM) and Nijmegenbreakage syndrome 1 (NBS1). However,genetic susceptibility to radiogenic cancer andother late effects of radiation exposure may bedetermined predominantly by co-inheritance of low-penetrance risk alleles, and how theseinteract with each other (gene-gene interac-

tions) and with radiation dose (gene-exposureinteractions). The development of high-densitypolymorphism arrays represents a promisingapproach in the search for genetic risk allelesconferring susceptibility to radiogenic cancer.

In addition to host factors and inherent geneticsusceptibility, there is evidence to suggest that the phenotype of the putative target cellfor transformation can impact on the risk ofdeveloping cancer after radiation exposure. For example, cells actively proliferating at thetime of exposure are predicted to be more susceptible to the adverse effects of radiation.In support of this, experimental evidencedemonstrates that dividing cells are more likelyto fix deoxyribonucleic acid (DNA) damage into mutation than nondividing cells and are,therefore, more susceptible to transformation.In some tissues, such as the breast, cellularproliferation is inversely correlated with age. As such, we might predict that radiogenic cancer risk would be higher in younger premenopausal women than older or post-menopausal women. Indeed, this seems to bethe case. Data such as these suggest that thepathological response to radiation-inducedDNA damage at the time of exposure, specifi-cally the balance between mutation and death

3:35 pm Patient Susceptibility to Radiation-Induced Cancer and Second Cancers Following Radiotherapy Procedures James M. AllanUniversity of York, UK


31

at the cellular level, might ultimately determinerisk of transformation. However, this modelremains to be challenged.

In summary, patient susceptibility to radiation-induced cancer is likely to be determined byinteracting genotypic and phenotypic charac-teristics. Despite its apparent complexity, an understanding of susceptibility to

radiotherapy-induced cancers could lead totherapeutic benefit such that patients at highrisk could be identified. Moreover, it is envis-aged that a focus on understanding the factorsthat predispose to the development of radio-therapy-induced cancers will also provide asound basis for the study of other late effectsin cancer survivors.

4:00 pm Panel DiscussionStephanie K. Carlson, Moderator

4:40 pm Closing RemarksThomas S. Tenforde, PresidentNational Council on Radiation Protection and Measurements

Mission Statement

32

Recognized worldwide as an authority on

radiation health protection for over 75 years.

The National Council on Radiation Protection and Measurements (NCRP) seeks to formulate and widely disseminate information, guidance and recom-mendations on radiation protection and measurements which represent theconsensus of leading scientific experts. The Council monitors areas in whichthe development and publication of NCRP materials can make an importantcontribution to the public interest.

The Council's mission also encompasses the responsibility to facilitate andstimulate cooperation among organizations concerned with the scientific andrelated aspects of radiation protection and measurements.

L.S. Taylor1929–1977

W.K. Sinclair1977–1991

C.B. Meinhold1991–2002

T.S.Tenforde2002–

Jerrold T. Bushberg, Chairman

Stephen Balter

James A. Brink

Cynthia C. Cardwell

Charles E. Chambers

Donald P. Frush

Linda A. Kroger

Edwin M. Leidholdt, Jr.

Fred A. Mettler, Jr.

Theodore L. Phillips

James E. Rodgers

J. Anthony Seibert

Thomas B. Shope, Jr.

Program Committee

33

2008 Annual Meeting

Low Dose and Low Dose-Rate Radiation Effects and Models

April 7-8, 2008

Arlington, Virginia

Registration

Monday, April 16, 2007 7:00 am – 5:00 pm

Tuesday, April 17, 2007 7:00 am – 1:00 pm

(No registration fee)

Register online (http://registration.ncrponline.org)

NCRP Publications(http://NCRPpublications.org)

Radiation Protection in Medicine

34

Publication Title Price ($)

Report No. 151 Structural Shielding Design and Evaluation for Megavoltage X- and Gamma-Ray Radiotherapy Facilities 100.00

Report No. 149 A Guide to Mammography and Other Breast Imaging Procedures 110.00

Report No. 147 Structural Shielding Design for Medical X-Ray Imaging Facilities 100.00

Report No. 140 Exposure Criteria for Medical Diagnostic Ultrasound; II. Criteria Based on All Known Mechanisms 95.00

Report No. 133 Radiation Protection for Procedures Performed Outside the Radiology Department 30.00

Report No. 107 Implementation of the Principle of As Low As Reasonably Achievable (ALARA) for Medical and Dental Personnel 35.00

Report No. 100 Exposure of the U.S. Population from Diagnostic Medical Radiation 40.00

Report No. 99 Quality Assurance for Diagnostic Imaging 50.00

Report No. 68 Radiation Protection in Pediatric Radiology 40.00

Commentary No. 9 Considerations Regarding the Unintended Radiation Exposure of the Embryo, Fetus or Nursing Child 20.00

Commentary No. 7 Misadministration of Radioactive Material in Medicine — Scientific Background 25.00

Excerpts from recent reviews of NCRP reports:

“This report [NCRP Report No. 151] has been long awaited by the therapy community and it servesat least two distinct communities of physicists: those newly entering the field that do not have alibrary shelf full of previous NCRP reports and the other group are the more experienced physiciststhat have all of the previous reports.”

J.B. Smathers[published in Journal of Applied Clinical Medical Physics, 7 (2006) 100–101]

“In conclusion NCRP Report 147 is well written and easily readable, and provides reference data in amanner that is easy to follow.”

G.J. Chalmers[published in Physics in Medicine and Biology, 50 (2005) 4243–4244]

Reports and commentaries are available from the NCRP website, http://NCRPpublications.org,in both soft- and hardcopy formats. Complete book

reviews of recent NCRP publications are also available at this website.

Contracts/Grants/Contributors/Sponsors

These organizations have supported the work of the National Council on RadiationProtection and Measurements during the period of January 1, 2006 to December 31, 2006.

ContractsDefense Threat Reduction AgencyU.S. Navy

GrantsCenters for Disease Control and PreventionNational Aeronautics and Space AdministrationNational Cancer InstituteNational Institute of Standards and TechnologyU.S. Department of EnergyU.S. Nuclear Regulatory Commission

ContributorsAmerican Academy of Health PhysicsAmerican Academy of Oral and Maxillofacial RadiologyAmerican Association of Physicists in MedicineAmerican College of Medical PhysicsAmerican College of Radiology FoundationAmerican Industrial Hygiene AssociationAmerican Nuclear SocietyAmerican Osteopathic College of RadiologyAmerican Roentgen Ray SocietyAmerican Society for Therapeutic Radiology and OncologyAmerican Society of Radiologic TechnologistsCouncil on Radionuclides and RadiopharmaceuticalsHealth Physics SocietyLandauer, Inc.Radiological Society of North AmericaSociety for Pediatric RadiologySociety of Nuclear Medicine

Corporate Sponsors3MDuke Energy CorporationGE HealthcareGlobal Dosimetry Solutions, Inc.Landauer, Inc.Nuclear Energy Institute

7910

Woo

dmon

t A

venu

e, S

uite

400

Bet

hesd

a, M

D 2

0814

-309

5ht

tp:/

/NC

RPo

nlin

e.or

ght

tp:/

/NC

RPp

ublic

atio

ns.o

rgD

A S

chau

er, E

xecu

tive

Dir

ecto

r

Advances in Radiation Protection in Medicine · medicine procedures used in cardiology and other...

Documents

Transcript of Advances in Radiation Protection in Medicine · medicine procedures used in cardiology and other...